Integral Ad Science (IAS) has been testing an ad verification program for connected TV, but it’s running into headwinds in a fragmented and unstandardized space.
The digital advertising ecosystem is notoriously replete with fraud. As television comes online, connected TVs (CTV) are now open to similar fraudulent activity, such as app-based fraud and device spoofing.
The more mature digital landscape, however, has a clearer set of standards that make it easier to measure fraudulent behavior. The relatively nascent playground of CTV isn’t afforded that same safety net.
In July, AdLedger found that 18% of ad requests for over-the-top inventory were fraudulent.
Sarah Canerday, business development director at IAS, said the company is bound by IAB and MRC guidelines on measuring invalid traffic. While a helpful starting point, Canerday said the rapidly growing CTV space is outpacing standards bodies.
“Because we are bound by these current MRC guidelines and checking these IAB lists, [and] because these publishers are quickly working to develop these apps, there are currently no standards governing the CTV app space, and it's wildly fragmented,” said Canerday. “Even across a single publisher they have different ways of serving ads into their CTV apps across their different properties.”
Canerday gave one example of a trusted device manufacturer in Australia that released a new streaming service, which was first marked as fraudulent. The service's user agent, a string that identifies an entity making an ad request, was not registered with the IAB, so traffic running through it was initially categorized as invalid.
Canerday said players in the relatively mature US CTV market understand how to get off the ground, but the wealth of devices and distributors can cause confusion post-launch.
“Even across a single publisher, [there can be] different ways of serving ads into their CTV apps across different properties,” said Canerday. “[Publishers] have different user agents that are not standardized. Because of all that mess... it needs a clean up.”
For example, an ad request from Pluto TV through Roku’s platform looks different than a request through Amazon Fire TV, Apple TV and Android TV. It also looks different than requests from smart TV manufacturers such as LG and Samsung.
After running its beta test, which measures ads from Verizon and around 15 other advertisers running across NBC Universal, CBS and a host of other publishers, IAS has three key recommendations to streamline CTV advertising.
First is supply-side data standardization. Getting publishers to support CTV-specific standardized data point would help separate fraud from the false positives caused by a lack of data, IAS says.
Second is hastening device registration. CTV device makers should prioritize reporting their user agent string with the IAB, IAS says.
Lastly is for publishers to standardize their identification practices. The IAB is currently working to standardize app IDs in CTV, but publishers in the meantime should be consistent and descriptive in how they name their properties, IAS says.
These would be first steps to help differentiate legitimate activity from fraud, which should help IAS tidy up the open exchange, Canerday said.
“When we talk to our advertisers and even our [demand-side platform] partners, people are pretty aligned with knowing that the things that you're going to find in the open exchange are probably the things you don't want your ads to be on…,” said Canerday, adding that private marketplace deals aren't always programmatic safehavens.
“Some PMPs out there that are bundles of CTV inventory across publishers. You're not working with a one-to-one direct publisher there, you're still working in some sort of marketplace, and that is potentially where we're going to uncover more of the longer-tail publishers that really do need to be validated as truly having CTV inventory.”
IAS will make its verification product generally available in the first quarter of 2020, and it’s next major effort will be to tackle brand safety in CTV.
One major dilemma is that publishers rarely share content-level information, meaning advertisers don’t often know where their ads appear.
Canerday said advertisers have to settle for app-level data, and that “money speaks” when pushing for more transparency.
“We want to make sure that they're aware of this issue, and that they're vocal with their media partners that they need content-level information,” said Canerday.